Nuclear imaging techniques are widely used as a research and clinical tool for imaging cancer and a variety of neurological and cardiovascular disorders. Improving diagnostics tools and monitoring new therapeutics for these diseases continues to pose challenges to existing imaging techniques and stimulates new directions in imaging instrumentation. Research to improve the performance of current nuclear medicine systems is very active (annual investment of several hundred million dollars) and outcomes need to be rapidly transferred to clinical devices to positively impact healthcare. Monte Carlo simulations are extensively used by scientists to design new nuclear medicine detectors and predict their performance. However value of these convenient simulation tools is largely dependent on the reliability of the data they produce. Accurate modeling of the major detector components is required to properly predict performance, thus avoiding costly design mistakes and expedite the implementation of novel detector designs into a scanner in the fastest and most cost-effective manner. While there have been tremendous improvements in the accuracy and flexibility of nuclear medicine simulation tools, they have one huge deficiency because they do not accurately describe light transport in the scintillator. Since the amount of scintillation lght, its spatial distribution and its temporal distribution ultimately define the spatial, energy and timing resolution of detectors, this is a huge limitation that has severely restricted confidence i simulation results. We have developed a new model of light transport in scintillators for nuclear medicine detector simulations that addresses the limitations of existing methods and propose to implement it in the widely-used simulation software GATE. This new approach differs from previous techniques by describing scintillator surfaces using 3D measurements instead of a simple analytical model. This work will be carried out by completing two specific aims and will build on our preliminary work on light transport modeling. First, we will develop and implement algorithms to compute reflectance properties of measured scintillator surfaces to be incorporated into a model of light transport in the existing GATE code. Second, we will validate the model using benchtop and commercial detectors. We anticipate that these developments will greatly improve the accuracy and reliability of optical simulations for nuclear imaging detectors. By rapidly disseminating our algorithms to the international scientific community, we will facilitate innovative instrumentation development and help accelerate the transition from detector concepts to implementation in nuclear medicine scanners in academic laboratories and companies across the world.